summary.mvr {pls.pcr} | R Documentation |
The information in the MVR object is summarised: the
function gives the dimensions of X and Y, the method used, and the
number of latent variables considered. If what
equals "all" or
"training", the RMS and R2 values of the training set are given; if
what
equals "all" or "validation", RMS and R2 values for the
cross-validation are given. In the latter case, a suggestion is made
about the optimal number of latent variables, as suggested in Hastie
et al. (2001).
summary.mvr(object, what, digits=4, print.gap=3, ...)
object |
An MVR object. |
what |
one of "all", "training" or "validation". |
digits |
Number of digits, default is 4. |
print.gap |
gap between columns of the print table. |
... |
other arguments to print.table . |
T. Hastie, R. Tibshirani and J. Friedman, The elements of statistical learning, Springer (2001).
data(NIR) attach(NIR) nir.mvr <- mvr(Xtrain, Ytrain, 2:8, valid="CV", grps=1) summary(nir.mvr)